adding copula param checks

Took 1 hour 3 minutes

sklarpy/copulas/_prefit_dists.py

@@ -118,6 +118,33 @@ def _get_mdists(self, mdists: Union[MarginalFitter, dict], d: int,
                                 "MarginalFitter object")
         return mdists
 
+    def _get_copula_params(self, copula_params: Union[Params, tuple],
+                           data_array: np.ndarray) -> tuple:
+        """Converts the user's copula-params input into tuple form and then
+        checks the parameters of the model and raises an error if one or more
+        is invalid.
+
+        Parameters
+        ----------
+        copula_params: Union[Params, tuple]
+            The parameters of the multivariate distribution used to specify
+            your copula distribution. Can be a Params object of the specific
+            multivariate distribution or a tuple containing these parameters
+            in the correct order.
+        data_array: np.ndarray
+            numpy array of the multivariate data.
+
+        Returns
+        -------
+        copula_params_tuple: tuple
+            The parameters which define the multivariate distribution in the
+            copula model.
+        """
+        params_tuple: tuple = self._mv_object._get_params(params=copula_params)
+        self._mv_object._check_dim(data=data_array, params=params_tuple)
+        return copula_params if isinstance(copula_params, tuple) \
+            else copula_params.to_tuple
+
     def __mdist_calcs(self, func_strs: List[str], data: np.ndarray,
                       mdists: Union[MarginalFitter, dict], check: bool,
                       funcs_kwargs: dict = None) -> Dict[str, np.ndarray]:
@@ -213,6 +240,10 @@ def logpdf(self, x: Union[pd.DataFrame, np.ndarray],
         # checking data
         x_array: np.ndarray = self._get_data_array(data=x, is_u=False)
 
+        # checking copula params
+        copula_params_tuple: tuple = self._get_copula_params(
+            copula_params=copula_params, data_array=x_array)
+
         # getting non-nan data
         mask, masked_data, output = get_mask(data=x_array)
 
@@ -225,7 +256,7 @@ def logpdf(self, x: Union[pd.DataFrame, np.ndarray],
 
         # calculating logpdf values
         logpdf_values: np.ndarray = self.copula_logpdf(
-            u=res['cdf'], copula_params=copula_params,
+            u=res['cdf'], copula_params=copula_params_tuple,
             match_datatype=False, **kwargs) + res['logpdf'].sum(axis=1)
 
         # converting to correct output datatype
@@ -319,6 +350,10 @@ def __cdf_mccdf(self, mc_cdf: bool, x: Union[pd.DataFrame, np.ndarray],
         # checking data
         x_array: np.ndarray = self._get_data_array(data=x, is_u=False)
 
+        # checking copula params
+        copula_params_tuple: tuple = self._get_copula_params(
+            copula_params=copula_params, data_array=x_array)
+
         # getting non-nan data
         mask, masked_data, output = get_mask(data=x_array)
 
@@ -330,7 +365,7 @@ def __cdf_mccdf(self, mc_cdf: bool, x: Union[pd.DataFrame, np.ndarray],
         mc_str: str = "mc_" if mc_cdf else ""
         func: Callable = eval(f"self.copula_{mc_str}cdf")
         copula_cdf_values: np.ndarray = func(
-            u=res['cdf'], copula_params=copula_params,
+            u=res['cdf'], copula_params=copula_params_tuple,
             match_datatype=False, **kwargs)
 
         # converting to correct output datatype
@@ -514,12 +549,17 @@ def copula_logpdf(self, u: Union[pd.DataFrame, np.ndarray],
         # checking data
         u_array: np.ndarray = self._get_data_array(data=u, is_u=True)
 
+        # checking copula params
+        copula_params_tuple: tuple = self._get_copula_params(
+            copula_params=copula_params, data_array=u_array)
+
         # calculating copula logpdf
-        g: np.ndarray = self._u_to_g(u=u_array, copula_params=copula_params)
+        g: np.ndarray = self._u_to_g(u=u_array,
+                                     copula_params=copula_params_tuple)
         g_logpdf: np.ndarray = self._mv_object.logpdf(
-            x=g, params=copula_params, match_datatype=False, **kwargs)
+            x=g, params=copula_params_tuple, match_datatype=False, **kwargs)
         copula_logpdf_values: np.ndarray = g_logpdf - self._h_logpdf_sum(
-            g=g, copula_params=copula_params)
+            g=g, copula_params=copula_params_tuple)
         return TypeKeeper(u).type_keep_from_1d_array(
             array=copula_logpdf_values, match_datatype=match_datatype,
             col_name=['copula_logpdf'])
@@ -601,11 +641,15 @@ def __copula_cdf_mccdf(self, mc_cdf: bool,
         # checking data
         u_array: np.ndarray = self._get_data_array(data=u, is_u=True)
 
+        # checking copula params
+        copula_params_tuple: tuple = self._get_copula_params(
+            copula_params=copula_params, data_array=u_array)
+
         # calculating cdf values
-        g: np.ndarray = self._u_to_g(u_array, copula_params)
+        g: np.ndarray = self._u_to_g(u_array, copula_params_tuple)
         mc_str: str = "mc_" if mc_cdf else ""
         func: Callable = eval(f"self._mv_object.{mc_str}cdf")
-        copula_cdf_values: np.ndarray = func(x=g, params=copula_params,
+        copula_cdf_values: np.ndarray = func(x=g, params=copula_params_tuple,
                                              match_datatype=False, **kwargs)
         return TypeKeeper(u).type_keep_from_1d_array(
             array=copula_cdf_values, match_datatype=match_datatype,